Background on ExpertNet
ExpertNet is an app designed to connect people with real experts. Often experts are expensive and inaccessible. The goal was of the app was to try and bridge this gap in a newand intuitive way.
Wait, but aren’t there plenty of services for
this already?
Yes, but upon deciding that I would make this type
of app I found many of them unpleasant to use or outdated. They didn’t seem user focused and many didn’t seem to fill much of a void that Youtube or Google Search didn’t already fill.
User Interviews
I found the answer of what was missing from the apps I initially studied by doing competitive analysis of other expert apps and 30 minute user interviews in the discovery portion of my research.
While conducting the interviews I took extensive notes which I then sorted into three separate categories.I also assigned each person a color of sticky note so I could match up responses and find patterns.
Features: Thoughts on existing features from competitors as well as desired features for my own.
Barriers: Things that get in the way of finding an expert through a service or a network.
Trust: I wanted to understand how and why we make that decision to trust someone as an expert.
“Expert” Definitions Vary
Over 73% of those I interviewed said they would trust someone who could show experience in a skill rather than a degree. While to others an expert is only the very best of the best.
Trust as a barrier
The biggest barrier to hiring an expert is trust. Existing apps show where to look, but don’t do a very good job in establishing trust with users. I would need to solve this problem in my design.
Features Needed
ExpertNet would need to be a place where an expert could not just post contact info, but actually show expertise through pictures and video.
Creating Personas
The following personas were created and based on the real insights and data from surveys and interviews conducted previously.
The data gathered helped me move one step closer to solving the dilemma of making a quality application to both find and communicate with experts.
Sergei is my prime persona and most features of the app are designed around him. He seeks experts for reassurance that he’s doing something correctly and is excited to to share what he knows with others and form a community.
“ The best tips are those shared face-to-face with another person.”
He’s most likely to create an Expert account himself after having his problem solved
Izzy has a high bar for what she considers to be an expert. Not just someone who has a skill, but somebody who is at the top of their field like she is.
“ I’m sure I can figure out my problem myself. If I get stuck I will just do more research”
While Izzy might not use all the features in the app, a few were designed around her such as the “Ask an Expert” feature. This feature allows her to ask her question as quickly as possible so she can determine whether the experts available meet her high standards.
First Iteration:
Wireframes & Paper Prototypes
Supercharged with the knowledge gained from my interviews and with the focus provided by my personas helped my first iteration be as productive as possible.
Thanks to UX and design thinking, the entire project shifted before I even started putting pencil to paper. If I would have gone about this normally. My first prototype was a very simple one wireframed on paper and grid paper. I mapped out the type of features I would need as well as the minimal steps it would take within the app to complete them.
Second Iteration:
Low to Mid Resolution
After creating and testing my paper prototype and gathering feedback. I began to do a more detailed mock up based on what I had learned. These low to mid resolution mock-ups where done in Balsamiq.
The following are showing the flow of the intended use of the app.
Responsive Design
My designs are responsive so it will not only look great on desktop as well as mobile. The images on the right depict the chat feature portion of the site.
More Testing
Card Sorting
Before moving onto a higher resolution prototype I conducted further testing. I performed card sorting through OptimalSort. I found the Similarity Matrix to be very useful. It confirmed that users thought a majority of my site was where they expected it to be. There was quite a bit to learn however and it lead to some design changes with the site map and layout. Luckily we found all of this before we went to high resolution mock ups. UX is amazing.
- 90% preferred to use a direct search method above other methods. This was reflected in the final design.
- 100% indicated that a video call was the preferred method to contact an expert once they had made a decision to do so.
Rainbow Sheet
I conducted another round of user interviews with iterative changes from the information I gained from the card sort. These interviews were guided and I encouraged my participants to think out loud. I recorded the calls and wrote down things that were said both positive and negative. Afterwards I sorted all comments and rated the changes that needed to be made by severity.
Changes to Sitemap
Based on the data from my guided interviews I made adjustments to the site structure. On the left I show before and after the edits.
What changed?
- Created a FAQ Section
- I moved the “Ask an Expert section” from it’s own category to be part of the “Finding an Expert” category because of the strong correlation between the two in the card sort.
- Changed the title of “Knowledge Sharing” to Community Forum based on similar titles from the card sort
Changes were made to the prototype based on another round of user interviews. I gave issues a severity rating and came up with a plan to deal with each one based on how much it affected the final outcome of the app.
Third Iteration:
High Resolution B & W
While I have moved onto High Resolution mock ups, I kept color out of all the elements except images. This helped users focus on my UI and not color preferences during testing. I’m glad I did!
Fourth Iteration:
Color & Material Design
It took awhile to get here, but I’m glad I took the steps I did. I’m miles ahead of where I would be if I just jumped to high resolution. I designed the screens to comply with Material Design standards.
So, What’s Next?
Hypotheses
The current version of the prototype lacks much of the social sharing and learning aspect I originally envisioned.
I plan to further build out the forum aspect of the prototype application. It’s currently very minimal. I predict that doing so will help the app feel more well rounded and to have a sense of community rather than just a marketplace to pay for advice.
How will I test?
I plan to conduct more user tests with specific prompts to see if users would turn to that specific function of the application to solve their problem. It’s very difficult to simulate how users would use a forum. At a certain point, the application needs to be built to have that functionality fully tested.
Once functioning, a beta forum could be tested with early access users. This would evolve testing.